simple question using dictionary...

[hidrkannan]

aDict = {}
aDict['a'] = 5
aDict['bc'] = 'st'
aDict['zip'] = 12345
 
 
 
def outputMethod(**aDict):
    print aDict['a']
    print aDict['bc']
    print aDict['zip']
    return
def outputMethod1(**aDict):
    for key in aDict:
        ...
        ...
    print a
    print bc
    print zip
    return
 
 
outputMethod(**aDict)
outputMethod1(**aDict)
 

with reference to the above code, I would like to define variables on the fly in outputMethod1 using the dictionary keys as variable names, so that i can print using just the variable names instead using the Dictionary keys as shown in ouputmethod. I am trying a for loop to extract the keys from the dictionary...but not sure how to define the "keys" as variable names?

Thanks in advance.

SKN

[woooee]

You can store a variable in a dictionary
a=5
test_dict["a"]=a
print test_dict["a"] prints 5
You can instead store the data in the dictionary since test_dict["a"] points to a points to 5, it is just as easy to store the value 5 as test_dict["a"]

You can also do test_dict[a]=3 and print a, test_dict[a] would print 5 3
or b = test_dict["a"], print b would print 5

Python uses pointers so you can pretty much store any Python object.

[hidrkannan]

You can store a variable in a dictionary
a=5
test_dict["a"]=a
print test_dict["a"] prints 5
You can instead store the data in the dictionary since test_dict["a"] points to a points to 5, it is just as easy to store the value 5 as test_dict["a"]

You can also do test_dict[a]=3 and print a, test_dict[a] would print 5 3
or b = test_dict["a"], print b would print 5

Python uses pointers so you can pretty much store any Python object.

Actually I would like to define variable names using the dictionary keys..to assign the corresponding dictionary values...

SKN

[Elias Alhanatis]

Hello!

I am a newbee too , so here is a simple solution i suggest:

aDict={'w':1,'e':2,'r':3}
index=0
for key in aDict.keys():
    exec(str(aDict.keys()[index])+'='+str(aDict[key]))
    index+=1
 

Hope i helped.....

[Elias Alhanatis]

Rushed to post the reply but forgot the case of values being strings!!!!!
Here is the improved ( yet still simple or 'naive' ) suggestion:

a={'w':1,'e':'rt6','r':3}
index=0
for key in a.keys():
    if type(a[key])!=type(""):
        exec(str(a.keys()[index])+'='+str(a[key]))
    else:
        exec(str(a.keys()[index])+'="'+str(a[key])+'"')
    index+=1
 
 

That will do the job!!!

Elias

[hidrkannan]

Rushed to post the reply but forgot the case of values being strings!!!!!
Here is the improved ( yet still simple or 'naive' ) suggestion:

a={'w':1,'e':'rt6','r':3}
index=0
for key in a.keys():
    if type(a[key])!=type(""):
        exec(str(a.keys()[index])+'='+str(a[key]))
    else:
        exec(str(a.keys()[index])+'="'+str(a[key])+'"')
    index+=1
 
 

That will do the job!!!

Elias

Thanks Elias. But I don't understand the use of index in the above code and also the if construct. how to retain the type of the value..say if the value of one of the keys of type list...

aDict = {}
aDict['abc'] = 1
aDict['defi'] = [2,4]
aDict['a'] = 5
aDict['bc'] = 'st'
aDict['zip'] = 12345
for kw in aDict:
    #exec(str(a.keys()[index])+'='+str(a[key]))
    exec(str(kw)+'="'+str(aDict[kw])+'"')
 
print a, bc,zip,abc, defi
 

Please comment.

SKN

[Elias Alhanatis]

Hi again!

The exec() function takes a string as an argument and just "makes it happen" ,
so all we must do is create a string that would give us the result we need if we
typed it in the interactive prompt.
The trick is that if a value in the dict IS a string , we must 'work our way around' in order not to confuse it with the exec() string we want to produce.
In any other case ( i suppose ) , there is no problem. ( you can confirm this by changing a value in my code to a list , examp: "w":[3,4,5] ).
In any case , to make your code error-proof , i sugest that you check the type of each value using the type() function. This is what i've done in my code ( line 4 ).

The index is used in order to iterate over the keys one by one , since we dont know ahead of time the number of keys in the dict , so we use it as a 'pointer' for each key.

Try to experiment with the code , google each word or function you don't understand and soon you'll get the idea.....
( Thats what i do.....)

I hope i helped a little!

Elias

[hidrkannan]

Hi again!

The exec() function takes a string as an argument and just "makes it happen" ,
so all we must do is create a string that would give us the result we need if we
typed it in the interactive prompt.
The trick is that if a value in the dict IS a string , we must 'work our way around' in order not to confuse it with the exec() string we want to produce.
In any other case ( i suppose ) , there is no problem. ( you can confirm this by changing a value in my code to a list , examp: "w":[3,4,5] ).
In any case , to make your code error-proof , i sugest that you check the type of each value using the type() function. This is what i've done in my code ( line 4 ).

The index is used in order to iterate over the keys one by one , since we dont know ahead of time the number of keys in the dict , so we use it as a 'pointer' for each key.

Try to experiment with the code , google each word or function you don't understand and soon you'll get the idea.....
( Thats what i do.....)

I hope i helped a little!

Elias

Thanks for your help. Still I am not convinced to use index for the above purpose. Basically I look for an "undict" function (as "dict" function puts all name=value pairs to a dictionary object) to divide the dict object to individual name=value pairs.
Somehow I feel that if else construct will reduce the performance.

Thanks again for the "exec" idea that I learnt today.

SKN

[Elias Alhanatis]

Inspired by the challenge , i will try to wrap it up in a
function without an 'if' clause and without an index. I will do it tonight , cause
right now i am at work ( LOL ) , so check this thread out later....

[bvdet]

I routinely import a dict object upon initializing a script and create the variables to avoid referencing the dictionary every time a value is required. This is the code:

for key, value in dd.items():
    exec "%s = %s" % (key, repr(value)) in None

You may not need the 'in None' in your application. In my applications, the values are limited to int, float, list, and string, and they are created correctly.

[hidrkannan]

I routinely import a dict object upon initializing a script and create the variables to avoid referencing the dictionary every time a value is required. This is the code:

for key, value in dd.items():
    exec "%s = %s" % (key, repr(value)) in None

You may not need the 'in None' in your application. In my applications, the values are limited to int, float, list, and string, and they are created correctly.

Thanks BV. Is it possible to have this as a general function and use it sort of an "undict" opposite of "dict" function...not sure how to return the equivalent "name = Value" pair statements to the calling method. Essentially I am looking for:

def aMethod(self,**aDict):
undict(aDict)
.....
.....

SKN

[hidrkannan]

Thanks BV. Is it possible to have this as a general function and use it sort of an "undict" opposite of "dict" function...not sure how to return the equivalent "name = Value" pair statements to the calling method. Essentially I am looking for:

def aMethod(self,**aDict):
undict(aDict)
.....
.....

SKN

BV,

Also what is the purpose of "in None"...It gave me setitem attribute error..but when I remove "in None", it worked !!

Thanks

SKN

[bvdet]

Thanks BV. Is it possible to have this as a general function and use it sort of an "undict" opposite of "dict" function...not sure how to return the equivalent "name = Value" pair statements to the calling method. Essentially I am looking for:

def aMethod(self,**aDict):
undict(aDict)
.....
.....

SKN

Yes. Assuming you want your undict() function in a module:

# module unDict
def unDict(dd, dd0):
    for key, value in dd.items():
        exec "%s = %s" % (key, repr(value)) in dd0

Pass the globals() dictionary to the function:

import unDict
 
dd = {'A': 'A string', 'B': 123.456, 'C': [1,2,3,4,5,6], 'D': 456}
unDict.unDict(dd, globals())
print A, B, C, D

Output:
>>> A string 123.456 [1, 2, 3, 4, 5, 6] 456

[bvdet]

BV,

Also what is the purpose of "in None"...It gave me setitem attribute error..but when I remove "in None", it worked !!

Thanks

SKN

An unqualified exec is not allowed in my applications. "in globals()" would work also.

[hidrkannan]

Thanks BV for the clarification. I need to understand the use of globals()...
Thanks again..

SKN

[hidrkannan]

Yes. Assuming you want your undict() function in a module:

# module unDict
def unDict(dd, dd0):
    for key, value in dd.items():
        exec "%s = %s" % (key, repr(value)) in dd0

Pass the globals() dictionary to the function:

import unDict
 
dd = {'A': 'A string', 'B': 123.456, 'C': [1,2,3,4,5,6], 'D': 456}
unDict.unDict(dd, globals())
print A, B, C, D

Output:
>>> A string 123.456 [1, 2, 3, 4, 5, 6] 456

Is there any other alternative to do this instead of using exec. I am hearing the use of exec is not a good practice as it is a dowside factor for performance?

Thanks in Advance
SKN

[bvdet]

Is there any other alternative to do this instead of using exec. I am hearing the use of exec is not a good practice as it is a dowside factor for performance?

Thanks in Advance
SKN

globals().update(dd)

The keys in dictionary dd are now available as variable names.

[hidrkannan]

Thanks BV.

I suppose, if I use with in a method/function, then I can use locals.update(dd) instead of globals().update(dd).
Please correct me if I am wrong.

SKN

[bvdet]

Thanks BV.

I suppose, if I use with in a method/function, then I can use locals.update(dd) instead of globals().update(dd).
Please correct me if I am wrong.

SKN

You cannot update the local dictionary - it's not allowed. It can be done by passing the global dictionary to the function for updating.

>>> dd = {'a': 1, 'b': 2}
>>> a
Traceback (most recent call last):
  File "<interactive input>", line 1, in ?
NameError: name 'a' is not defined
>>> def update_dict(dd_to_update, dd):
...     dd_to_update.update(dd)
...     
>>> update_dict(globals(), dd)
>>> a
1
>>> 

[hidrkannan]

You cannot update the local dictionary - it's not allowed. It can be done by passing the global dictionary to the function for updating.

>>> dd = {'a': 1, 'b': 2}
>>> a
Traceback (most recent call last):
  File "<interactive input>", line 1, in ?
NameError: name 'a' is not defined
>>> def update_dict(dd_to_update, dd):
...     dd_to_update.update(dd)
...     
>>> update_dict(globals(), dd)
>>> a
1
>>> 

If I use globals().update(dd) then the variables become global. I would like it to keep it local.

[bvdet]

If I use globals().update(dd) then the variables become global. I would like it to keep it local.

It could be encapsulated in a module. The global namespace for a function is always the module in which it was defined.

>>> import amodule
>>> amodule.update_dict(amodule.update_dict.func_globals, {'value1': 10.5, 'value2': 127.6})
>>> amodule.value1
10.5
>>> 

[hidrkannan]

Thanks again BV.

SKN